The 2012 Monitoring the Future survey found that 6.5% of high school seniors smoke marijuana daily, up from 5.1% from five years ago. At the same time, only 44.1% of older teens see regular marijuana use as harmful (the lowest since 1979). Because perceived harm is a key indicator of use and movements to legalize marijuana (MJ) are succeeding in the US, these trends will continue and may lead to a new health crisis. While the debate as to the harms of MJ rages, neuroimaging studies have not realized their full potential to inform the national dialogue on this key social issue. For example, magnetic resonance imaging (MRI) studies suggest some differences in brain structure and function associated with MJ use, however, the findings are equivocal. One possible reason for this is that the effects of MJ use are so intertwined with specific drug use patterns in the individual that it becomes difficult to identify differences between MJ users and controls using conventional statistical methods (because there is such large variability in MJ users). To address this gap, this project combines MRI data from five different NIDA-funded studies of chronic MJ use to derive benefit from having a much larger sample of MJ users. A new statistical method suitable only with large datasets, called data fusion, will then be applied to the combined dataset. Data fusion capitalizes on the wide range of effects that MJ use may engender, rather than being confounded by it. A further strength of this method is that it integrates information across multiple MRI measurements to link together, for example, gray and white matter structure with circuit-level behavior of brain networks. Analyses of extant data will focus on investigating the effects of MJ use on executive function and cognitive control in MJ users by: 1) assessing interactions between large-scale brain networks that coordinate together during cognitive tasks, 2) applying the data fusion approach to link cognitive network function with brain structure and, in turn, disentangle the impact of chronic MJ use, and 3) most importantly, testing the predictive value of these findings in another large dataset that will be available in the third year of this study. Validating relationships between structure-function patterns and MJ use in a new sample is a significant advancement to addiction neuroimaging that will lead to novel biomarkers to target the detection of vulnerable individuals and to suggest new diagnostic, prevention and treatment strategies. Consistent with NIDA's strategic goals of treatment and prevention, the proposed research will: 1) Advance our understanding of the neurobiological effects of chronic marijuana use on brain structure and network circuitry, 2) Use novel statistical analyses of extant data combined with validation in a different dataset to identify biomarkers to characterize vulnerable individuals and suggest new diagnostic, prevention, and treatment strategies.